User and product embeddings

Mar 10, 2022
Open in Github

We calculate user and product embeddings based on the training set, and evaluate the results on the unseen test set. We will evaluate the results by plotting the user and product similarity versus the review score. The dataset is created in the Get_embeddings_from_dataset Notebook.

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from ast import literal_eval

df = pd.read_csv('data/fine_food_reviews_with_embeddings_1k.csv', index_col=0)  # note that you will need to generate this file to run the code below
df.head(2)
ProductId UserId Score Summary Text combined n_tokens embedding
0 B003XPF9BO A3R7JR3FMEBXQB 5 where does one start...and stop... with a tre... Wanted to save some to bring to my Chicago fam... Title: where does one start...and stop... wit... 52 [0.03599238395690918, -0.02116263099014759, -0...
297 B003VXHGPK A21VWSCGW7UUAR 4 Good, but not Wolfgang Puck good Honestly, I have to admit that I expected a li... Title: Good, but not Wolfgang Puck good; Conte... 178 [-0.07042013108730316, -0.03175969794392586, -... 
df['babbage_similarity'] = df["embedding"].apply(literal_eval).apply(np.array)
X_train, X_test, y_train, y_test = train_test_split(df, df.Score, test_size = 0.2, random_state=42)

user_embeddings = X_train.groupby('UserId').babbage_similarity.apply(np.mean)
prod_embeddings = X_train.groupby('ProductId').babbage_similarity.apply(np.mean)
len(user_embeddings), len(prod_embeddings)

(577, 706)

We can see that most of the users and products appear within the 50k examples only once.

2. Evaluate the embeddings

To evaluate the recommendations, we look at the similarity of the user and product embeddings amongst the reviews in the unseen test set. We calculate the cosine distance between the user and product embeddings, which gives us a similarity score between 0 and 1. We then normalize the scores to be evenly split between 0 and 1, by calculating the percentile of the similarity score amongst all predicted scores.

from utils.embeddings_utils import cosine_similarity

# evaluate embeddings as recommendations on X_test
def evaluate_single_match(row):
    user_id = row.UserId
    product_id = row.ProductId
    try:
        user_embedding = user_embeddings[user_id]
        product_embedding = prod_embeddings[product_id]
        similarity = cosine_similarity(user_embedding, product_embedding)
        return similarity
    except Exception as e:
        return np.nan

X_test['cosine_similarity'] = X_test.apply(evaluate_single_match, axis=1)
X_test['percentile_cosine_similarity'] = X_test.cosine_similarity.rank(pct=True)

import matplotlib.pyplot as plt
import statsmodels.api as sm


correlation = X_test[['percentile_cosine_similarity', 'Score']].corr().values[0,1]
print('Correlation between user & vector similarity percentile metric and review number of stars (score): %.2f%%' % (100*correlation))

# boxplot of cosine similarity for each score
X_test.boxplot(column='percentile_cosine_similarity', by='Score')
plt.title('')
plt.show()
plt.close()

Correlation between user & vector similarity percentile metric and review number of stars (score): 29.56%
image generated by notebook

We can observe a weak trend, showing that the higher the similarity score between the user and the product embedding, the higher the review score. Therefore, the user and product embeddings can weakly predict the review score - even before the user receives the product!

Because this signal works in a different way than the more commonly used collaborative filtering, it can act as an additional feature to slightly improve the performance on existing problems.